From the second half of last year to this year, AI chips have blossomed everywhere. Many domestic AI startups have launched their own AI-specific chips, but Victor Peng, president and CEO of Xilinx, the world's largest FPGA manufacturer, poured a cold water and thought that startups are not You should start your own AI-specific chip from scratch.
On October 16th, at the Xilinx Developers Conference, Victor told reporters that many start-ups have no funds to develop and mass produce AI chips because of the huge cost of research and development. AI startups should focus on innovative algorithms and architectures rather than designing chips. “How many startups have succeeded in doing ASICs (dedicated custom chips)?”
Advances in technology have made competition among different processor vendors such as CPUs, GPUs, FPGAs (Field Programmable Gate Arrays) and ASICs (Application Specific Integrated Circuits). As the leader of FPGAs, Victor said that Xilinx's competitors are no longer the second largest manufacturer of FPGAs, Altera (acquired by Intel), but the processor business of NVIDIA and Intel.
Due to its high flexibility, FPGAs are considered to be an intermediate solution when the AI algorithm is not mature. The biggest advantage is that the hardware functions of the system can be modified by software like software. Compared with GPU and CPU general-purpose chips, it has higher performance and lower energy consumption.
Xilinx believes that FPGA chips will be the highlight. In order to better meet the challenges of the AI era, the company further launched the first new category platform on the 16th - Versal ACAP (Adaptive Compute Acceleration Platform).
The Versal portfolio is based on TSMC's 7 nm FinFET process technology and is a platform that combines software programmability with hardware acceleration and flexibility in a specific area. Victor pointed out: "With the explosive development of AI and big data and the slowdown of Moore's Law, the industry has reached a critical turning point. The cycle of chip design has been unable to keep up with the pace of innovation. Versal supports all types of developers through optimization Software and hardware to speed up their overall application, while providing immediate flexibility."
Because of the AI vent, the stock price of Nvidia has risen sharply in the past two years. Just in February of this year, Google announced the opening of the TPU (tensor processor) service, joining the battle of AI chips. TPU is Google's custom chip for machine learning and is an ASIC.
In an interview with reporters from China Finance and Economics, Victor said that the AI chip market will not only have one kind of chip architecture, but it is not optimistic about dedicated chips.
Compared with CPU and GPU, the biggest advantage of FPGA is its highly adaptive strain capability. "The GPU does have its own advantages for some applications and workload acceleration. In the machine learning world, the GPU does integrate some new module templates to speed up machine learning, but its performance is fixed for a fixed period of time. FPGAs can be accelerated for different workloads, and they perform much better than GPUs in this area, and can be applied to different networks during machine learning."
For Chinese start-ups to build AI-specific chips, Victor bluntly, it is not a technical reason. Some insiders told reporters that some start-ups are mainly for financing.
"If you really let these companies create value in the high-tech field, you have to do things that others have not done, rather than doing things that several big companies are doing, which is a waste of resources and capital." Victor told reporters. It’s not that start-ups can’t be ASICs. “If you can do better than Intel, NVIDIA, and Xilinx, but more startups should focus on specific areas and applications, rather than developing chips from scratch, because A lot of companies are doing it."
On July 18, Xilinx announced the completion of the acquisition of Shenjian Technology, a domestic AI startup. As a domestically influential AI chip startup company, the acquisition of Shenjian Technology has made many people feel very sudden and unexpected. Some insiders also said that Shenjian Technology may have encountered bottlenecks in its business development. At the same time, this incident "let everyone see the fact that the chip is not that simple."
Salil Raje, Xilinx's vice president of software and IP products, pointed out in an exclusive interview with the first financial reporter that most AI chip companies will be eliminated. AI will change significantly in the next few years, so flexible hardware is needed. "AI-specific chips will be used in a specific vertical field, such as smart city camera chips, but not universal chips."
Accelerate integration with Shenjian Technology team
At the same time, Victor also made some responses to the newly acquired Chinese AI startup, Shen Jian Technology. He told the First Financial Reporter that Shen Jian Technology has not yet brought substantial income, but hopes that the team and Xilinx can complete the integration as soon as possible to promote the company's business growth in the Chinese market.
Founded in 2016, Shenjian Technology focuses on neural network pruning, deep compression technology and system-level optimization. In recent years, it has also been developing AI chips.
Victor said that since Shenjian Technology is mainly an engineering department, the company will report to Salil after the acquisition, and the market and other commercial departments of Shenjian will be taken over by Xilinx and further expanded. He said that Xilinx will help Shenjian Technology to serve customers around the world, as well as its technology, and commercial marketing.
Victor told reporters that the reason why they value the deep insights lies in their network optimization, DNN and some architecture and practical technology. However, because Shenjian Technology mainly focuses on R&D and does not have much income, it will not bring substantial income to Xilinx in the short term. However, Victor also said that through the cooperation after the merger, Xilinx will recommend Shenjian Technology's products to more customers and increase the latter's income, which in turn will enhance Xilinx's overall competitiveness.
After being acquired by Xilinx, Shenjian Technology will focus on the FPGA field and will not develop its own AI chip. Salil told reporters, "They realized that after the chip (ASIC) was released, it did not perform well on some new neural network layers, and the efficiency was very low, which is why we do not intend to let them develop ASIC chips."
